2D Metabolomics NMRPipe Processing

NMRPipe Processing to obtain .ft2 and .nv files

1. The data files from ICONNMR can be used directly by NMRPipe to process the 2D 1H–13C-HSQC spectra.

2. On a Linux workstation, open a terminal and go to the directory that contains the NMR data. Type bruker to start the NMRpipe software.

3. Read in the experimental parameters file by clicking Read Parameters and verify that all the parameters have been correctly updated. Confirm that the mode of data collection has been set to echo-antiecho if the NMR spectrum was collected with the hsqcetgpsisp2 pulse program.

4. Click Update Script to save an NMRPipe processing script fid.com file in the working directory.

5. Type ./fid.com to start the NMRPipe processing script.

6. When the NMRPipe processing has finished, type nmrDraw to view the processed NMR spectrum.

7. Phase the NMR spectrum in NMRpipe and note the p0 and p1 values for both the 1H and 13C dimensions.

8. Edit the NMRPipe processing script hsqcproc.com and replace the parameters associated with the NMRPipe phase correction command, ps, with the p0 and p1 values obtained from step 7.

9. Type ./hsqcproc.com to start the NMRPipe processing script.

10. Repeat steps 3–9 for each 2D 1H–13C-HSQC NMR spectrum in the dataset. This produces a set .ft2 files. One .ft2 file is created for each 2D 1H–13C-HSQC NMR spectrum collected for each replicate from each group.

11. Copy all of the .ft2 files into a new folder and use the NMRPipe script addnmr.com to generate NMRviewJ files from the .ft2 files. A .nv file will be generated for each individual spectrum (.ft2 file) with a numerically incremented root name of Final_. In addition, the script will combine all of the NMR spectra together into a single file called results.nv. The script will also generate the text file, rate.txt, which lists all of the individual. nv files (Final_).

---

Peak Picking and Peak Integration of 2D 1H–13C-HSQC Spectra in NMRviewJ

1. Type nmrviewj to start NMRviewJ.

2. From the Dataset toolbar in the main window, use the Open and Draw Datasets function to select the result.nv file.

3. Right click and select attributes to open the attributes window.

4. In the attributes window, select the PeakPick tab.

5. In the blank Lists field in the attribute window, type a filename (i.e., lists) for the new peak pick list. Click the Pick button. The software will automatically peak pick the displayed spectrum and populate lists with the peak ID number, chemical shifts, and intensity.

6. Choose Show Peak Table from the Peak toolbar on the main window. A peak table window will open that lists the peak ID, peak intensity, and the peak chemical shifts.

7. Manually edit the peak list and remove solvent peaks, noise peaks, or other spectral artifacts. Peaks are deleted from the peak table by using the delete function in the PeakPick tab in the attributes window along with the spectrum display window. In the spectrum display window, use the mouse to position the two cursors around any peak or spectral region to form a box. Then, click the Delete button under the PeakPick tab in the attributes window to remove the peak(s).

8. After the peak table has been completely edited, on the peak table window, choose the Edit tab and select Compress and Degap. Answer yes to the pop-up question. This will finalize changes to the peak list and prevent any further edits.

9. On the peak table window, choose the Edit tab and then select Save Table. A file browser window will open in order to choose a name and location to save the new peak list file. The saved peak pick file can be viewed and edited in Excel.

10. In order to obtain peak intensities across the entire set of NMR spectra in the dataset, click on the Analysis tab on the main window and select Rate Analysis. A setup window for the Rate Analysis will open.

11. In the Rate Analysis setup window:

(a) Set the Prefix for matrix numbers field to Final_.

(b) Set the Peaklist field to lists (defined in step 5).

(c) Make sure Auto fit is checked.

(d) Use all other default settings.

(e) Click Load time file.

(f) In the file browser window, select rate.txt (created in NMRPipe Processing, step 11).

(g) Click Measure All. The software automatically populates the table in the Rate Analysis setup window with all of the peak intensities across the entire NMR dataset.

(h) Click Save Table. In the file browser window, save the peak intensities table to a new filename (i.e., intensities).

12. The peak list (i.e., list) and the peak intensities (i.e., intensities) files are merged in Microsoft Excel using the common peak ID column. The ppm1 (1H ppm) and ppm2 (13C ppm) columns are added to the peak intensities columns to generate a complete matrix of NMR peaks and intensities across the entire dataset.

13. The merged Excel file is saved to a new filename.

---

Metabolite Assignments from 2D 1H–13C-HSQC Peak Lists

1. The complete list of peaks obtained from the NMRviewJ analyses is searched using NMR metabolomics databases such as HMDB, BMRB, or other databases.

2. On the HMDB homepage, choose the Search tab and select 2D NMR Search.

3. From the Spectra Library pull-down menu, choose 13C HSQC.

4. Cut and paste the 2D 1H–13C-HSQC peak lists into the Peak List field. One set of 1H and 13C chemical shifts, respectively,per line. Chemical shift values should only be separated by white space.

5. Set the 1H chemical shift error tolerance to 0.05 ppm (X-axis Peak Tolerance +/- field) and the 13C chemical shift error tolerance to 0.10 ppm (Y-axis Peak Tolerance +/- field).

6. Click the Search button. Depending on the size of the peak list,the software will return a ranked-order list of possible metabolites based on the number of chemical shift matches to reference spectrum.

7. Manually curate the list of potential metabolite assignments based on the number of chemical shift assignments, the quality of the spectral overlap (i.e., chemical shift match), the number of other metabolites in the same metabolic pathway, and the biological system (i.e., is it a reasonable or possible metabolitefor the organism).

8. Obtain additional NMR (e.g., HMBC, HSQC-TOCSY) and/or MS spectral data to confirm or refute the assignment.

9. An assigned 2D 1H–13C-HSQC spectrum